Pipeline

From Fmepedia

Pipeline is the term used to describe the flow of features between reader and writer, and therefore includes any transformation functions that are carried out on a feature.

The term can be used in a generic way ("my features passed through the pipeline") or in a more specific way to describe a file that contains a set of defined transformation functionality. In the latter sense a pipeline is similar to a Mapping File, which itself is a non-graphic version of a Workspace. The difference is that a pipeline doesn't have defined readers or writers, just a set of processing commands. The typical extension for a pipeline file is either ".fmi" or ".fme".

A workspace might be turned into a pipeline by first saving it as a mapping file (File > Save As) and then editing the mapping file in a text editor to delete the source and destination dataset definitions.

Pipelines can be used for...

• Processing data in the FME Viewer. Right click a 'View' entry in the navigation pane and select the option 'Apply Pipeline...'
• Processing data in a Gentrans translation. Use the pipeline keyword to define the pipeline file.
• Processing data as a branch in an existing translation. Define the pipeline using the INCLUDE keyword in your Mapping File or workspace header.